Report

The effect of spike mutations on SARS-CoV-2

neutralization
Graphical Abstract Authors
Chloe Rees-Spear, Luke Muir,
Sarah A. Griffith, ..., Katie J. Doores,
Marit J. van Gils, Laura E. McCoy

Correspondence
l.mccoy@ucl.ac.uk

In brief
This study describes neutralization by
antibodies and convalescent sera of
SARS-CoV-2 spike mutants. Rees-Spear
et al. show that SARS-CoV amino acid
substitutions and the B.1.1.7 variant can
block monoclonal antibody neutralization
and that serum samples collected
following mild illness are less resilient to
spike variation than those following
severe illness.

Highlights
d SARS-CoV-2 pseudotypes produced by amino acids
substitutions in SARS-CoV

d SARS-CoV-2 pseudotyped virus that encodes the B.1.1.7

variant spike

d Amino acid changes and B.1.1.7 can decrease monoclonal

antibody neutralization

d Minimal effect on sera with only 10% losing potency against

the B.1.1.7 pseudotype

Rees-Spear et al., 2021, Cell Reports 34, 108890

March 23, 2021 ª 2021 The Authors.
https://doi.org/10.1016/j.celrep.2021.108890 ll
 ll
OPEN ACCESS

Report
The effect of spike mutations
on SARS-CoV-2 neutralization
Chloe Rees-Spear,1 Luke Muir,1 Sarah A. Griffith,1 Judith Heaney,2 Yoann Aldon,3 Jonne L. Snitselaar,3 Peter Thomas,1
Carl Graham,4 Jeffrey Seow,4 Nayung Lee,1 Annachiara Rosa,5 Chloe Roustan,5 Catherine F. Houlihan,2,6
Rogier W. Sanders,3 Ravindra K. Gupta,7 Peter Cherepanov,5 Hans J. Stauss,1 Eleni Nastouli,2,5,8 on behalf of the SAFER
Investigators, Katie J. Doores,4 Marit J. van Gils,3 and Laura E. McCoy1,9,*
1Instituteof Immunity and Transplantation, Division of Infection and Immunity, University College London, London NW3 2PF, UK
2Advanced Pathogens Diagnostic Unit, Department of Clinical Virology, University College London Hospitals NHS Foundation Trust, London
W1T 4EU, UK
3Amsterdam University Medical Centers, Amsterdam Institute for Infection and Immunity, University of Amsterdam, 1105 AZ Amsterdam, the

Netherlands
4School of Immunology and Microbial Sciences, King’s College London, London SE1 9RT, UK
5The Francis Crick Institute, London NW1 1AT, UK
6Research Department of Infection, Division of Infection and Immunity, University College London, London WC1 6BT, UK
7Department of Medicine, University of Cambridge, Cambridge CB2 0AW, UK
8Great Ormond Street Institute for Child Health, Infection, Immunity and Inflammation, University College London, London WC1N 1EH, UK
9Lead contact

*Correspondence: l.mccoy@ucl.ac.uk
https://doi.org/10.1016/j.celrep.2021.108890

SUMMARY

Multiple severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) vaccines show protective efficacy,
which is most likely mediated by neutralizing antibodies recognizing the viral entry protein, spike. Because
new SARS-CoV-2 variants are emerging rapidly, as exemplified by the B.1.1.7, B.1.351, and P.1 lineages, it
is critical to understand whether antibody responses induced by infection with the original SARS-CoV-2 virus
or current vaccines remain effective. In this study, we evaluate neutralization of a series of mutated spike
pseudotypes based on divergence from SARS-CoV and then compare neutralization of the B.1.1.7 spike
pseudotype and individual mutations. Spike-specific monoclonal antibody neutralization is reduced dramat-
ically; in contrast, polyclonal antibodies from individuals infected in early 2020 remain active against most
mutated spike pseudotypes, but potency is reduced in a minority of samples. This work highlights that
changes in SARS-CoV-2 spike can alter neutralization sensitivity and underlines the need for effective real-
time monitoring of emerging mutations and their effect on vaccine efficacy.

INTRODUCTION (COVID-19) (McMahan et al., 2021; Polack et al., 2020). Howev-

Serum neutralization activity is a common correlate of protection
against viral infection following vaccination or natural infection
(Plotkin, 2008). However, effective protection from viral infection
can also require a sufficient breadth of serum neutralization
rather than potency alone. This is because of the high levels of
variation observed in major antigens across some viral popula-
tions (Burton et al., 2012). For example, in the response against
influenza, the majority of neutralizing serum antibodies target
only a particular set of influenza strains as a result of antigenic
drift of the immunodominant hemagglutinin head (Zost et al.,
2019). Because of this, an annual vaccine is required and must
be matched to the most probable circulating strain in any given
year to ensure protection from infection. Data emerging from hu-
man vaccine trials and challenge studies in animal models sug-
gest that neutralizing antibodies can prevent disease caused
by infection with severe acute respiratory syndrome coronavirus
2 (SARS-CoV-2), the virus that causes coronavirus disease 2019
er, new variants of SARS-CoV-2 have begun to emerge (Kemp
et al., 2020; Oude Munnink et al., 2021; Tegally et al., 2020;
Welkers et al., 2021). These variants include mutations in the ma-
jor neutralizing antigen, the spike glycoprotein, and raises the
question of whether neutralizing serum responses induced by
early circulating strains or by vaccines based on the spike
sequence of these early strains can neutralize the recently
emerged virus variants.
Prior to emergence of multiple mutations in spike in the human
population, we reasoned that a logical way to identify potential
escape mutations was to look at sites of amino acid variation
relative to the most closely related human betacoronavirus,
SARS-CoV, which caused the original SARS outbreak (CDC,
2003). These two closely related viruses are characterized by
notable differences in transmission dynamics and disease out-
comes (Cevik et al., 2020; Lipsitch et al., 2003; Petersen et al.,
2020), but use the human ACE2 protein as a viral entry receptor
(Li et al., 2003) and share approximately 75% similarity overall in

Cell Reports 34, 108890, March 23, 2021 ª 2021 The Authors. 1
This is an open access article under the CC BY license (http://creativecommons.org/licenses/by/4.0/).
 ll
OPEN ACCESS Report

(legend on next page)

2 Cell Reports 34, 108890, March 23, 2021


Report
spike at the amino acid level (Gralinski and Menachery, 2020).
Both viruses use the same region of their respective spikes to
bind ACE2, the receptor binding domain (RBD; found in the S1
subunit of spike). There is considerable amino acid variation be-
tween the two RBDs despite their conserved binding to ACE2,
which explains why the majority of COVID-19 sera have weaker
or no neutralizing activity against SARS-CoV, but cross-neutral-
izing monoclonal antibodies (mAbs) have been isolated (Brouwer
et al., 2020).
Since the start of the pandemic, sequencing of virus popula-
tions has been deployed to enable detection of individual muta-
tions in SARS-CoV-2. Recently, a new variant, B.1.1.7, has
emerged in the United Kingdom (Kemp et al., 2020; Rambaut
et al., 2020) that includes multiple mutations in the RBD and
the N-terminal domain (NTD) of spike, targets for neutralizing an-
tibodies. Similarly, additional variants have been identified in
South Africa (B.1.351) and Brazil (P.1) (Faria et al., 2021; Tegally
et al., 2020). The B.1.351 and P.1 variants share a deletion of
three amino acids in Orf1ab and key mutations in the RBD
(E484K and N501Y); data so far consistent with convergent evo-
lution and recombination (Varabyou et al., 2020). Early reports
indicated that, although the RBD mutation N501Y in the
B.1.1.7 strain does not compromise post-vaccine serum neutral-
ization (Xie et al., 2021), the additional changes in the B.1.351
strain do impair neutralization (Greaney et al., 2021; Wibmer
et al., 2021).
In this study, we evaluated the potential role of individual
amino acids in facilitating escape from neutralizing antibodies.
First, we made a series of point mutations to change the amino
acids in SARS-CoV-2 to those found at the analogous position
in SARS-CoV. Second, we made individual point mutations
emerging in real-world populations and generated a pseudotype
virus using the B.1.1.7 variant spike sequence. We identify mul-
tiple mutations that can abrogate neutralization by some mAbs
targeting the RBD of spike. However, in contrast, we show that
serum responses are more resilient to these mutations, espe-
cially following severe illness, where the antibody response is
characterized by increased breadth.
RESULTS
Generation of potential escape mutants by SARS-CoV
amino acid substitution
There are 56 individual amino acid changes between the RBD of
SARS-CoV-2 and SARS-CoV (Ortega et al., 2020). We prioritized
15 of the 56 changes by considering which mutations resulted in
amino acids of substantially different biochemical character and
which changes occurred in sequential positions. These sites
were mutated in the SARS-CoV-2 spike to match SARS-CoV
(Figure S1) and used to produce pseudotyped viruses (Seow
et al., 2020). Twelve of the 15 mutated pseudotypes gave virus
Figure 1. Mutating amino acids in SARS-CoV-2 spike to match SARS-CoV decreases mAb neutralization
(A) The indicated mAbs were assessed by pseudotyped neutralization assay. Data are representative of three independent repeats. The horizontal dotted line in
each graph indicates 50% neutralization.
(B) IC50 values for each mAb against the mutant SARS-CoV-2 pseudotyped viruses indicated in the left column. The previously established binding cluster and
binding to RBD for each mAb are indicated above each column.
See also Figure S1.
ll
OPEN ACCESS
titers and were then screened for any alteration in neutralization
against a panel of human mAbs (Brouwer et al., 2020) isolated af-
ter SARS-CoV-2 infection. These mAbs have been mapped pre-
viously into 11 binding clusters, where mAbs within a cluster
compete reciprocally for binding to spike. Representatives of
each neutralizing cluster were selected for evaluation against
the spike mutant pseudotypes.
Effect of SARS-CoV spike substitutions on SARS-CoV-2
mAb neutralization
Initial screening assays showed no major effect on neutralization
by any of the mAbs against pseudotypes encoding RFA346-8KFP,
S459G, and ST477-8GK. In contrast, the remaining nine viral pseu-
dotype mutants diminished neutralization for at least one mAb,
as described below (Figures 1A and 1B).
P384A
The P384A substitution resulted in complete loss of neutralization
by COVA1-16, a cluster III RBD-specific mAb that allosterically
competes with ACE2 rather than directly blocking the binding
site (Liu et al., 2020). This mutation has been described and char-
acterized structurally elsewhere (Wu et al., 2020), revealing that
this proline-to-alanine change results in a relatively small alter-
ation in protein structure that can enable SARS-CoV mAbs to
neutralize SARS-CoV-2 P384A. However, P384A does not weaken
neutralization by any other mAbs, including another mAb in the
cluster III competition group.
K417V
The K417V mutation results in a pseudotyped virus that is less
susceptible to COVA2-07-mediated RBD-specific neutraliza-
tion. That this mutation should affect this mAb, which competes
directly with ACE2 for binding, is not unexpected because the
lysine at position 417 forms a hydrogen bond with ACE2 (Lan
et al., 2020) that is likely disrupted by this substitution. We then
evaluated an additional mAb, COVA2-04, from the same
competitive binding cluster as COVA2-07. This is because
COVA2-04 is representative of a class of SARS-CoV-2-neutral-
izing antibodies that use the VH3-53 gene (Cao et al., 2020;
Mor et al., 2020; Robbiani et al., 2020). COVA2-04 was not
able to neutralize the K417V pseudotype (data not shown).
KVG444-6TST
This multiple substitution, which is a substantial change between
SARS-CoV-2 and SARS-CoV, results in a 3.7-fold drop in
neutralization potency for COVA2-29, which is a cluster I RBD-
specific antibody. This is the largest effect of this mutation; the
neutralization activity of the other mAbs is largely unaffected
despite alteration of three sequential amino acids. This may be
explained by the relatively minor differences in the amino acid
side chains at the mutated residues.
L452K
This mutation is situated directly in the receptor binding motif
(RBM) of the RBD. It renders pseudotyped virus resistant to
Cell Reports 34, 108890, March 23, 2021 3
 ll
OPEN ACCESS
neutralization by COVA2-29 but does not affect the other cluster I
mAb COVA1-18 or any other mAb tested.
LF455-6YL
This double substitution reduces neutralization by RBD-specific
mAbs from different clusters; specifically, the cluster I mAb
COVA2-29, cluster III mAb COVA2-07, and cluster VI mAb
COVA1-12. For COVA1-12, all neutralization activity is abol-
ished, whereas COVA2-07 activity is just below the level required
to calculate a 50% inhibitory concentration (IC50).
TEI470-2NVP
This triple mutation is located in a loop within the RBM where
other substitutions have been reported to abolish ACE2 binding
(Xu et al., 2021; Yi et al., 2020). This mutation prevents neutrali-
zation by COVA2-29 (cluster I), COVA2-07 (cluster III), and
COVA2-02 (cluster VII). It also reduces the activity of the most
potent mAb, COVA1-18 (cluster I), whereas this mAb is only mini-
mally affected by other mutations. Moreover, TEI470-2NVP lowers
the potency of the structurally unmapped non-RBD cluster XI
mAb, COVA1-21, to the limit of detection.
S494D
This single substitution toward the end of the RBM destroys
neutralization activity by COVA2-29 (cluster I) and COVA1-12
(cluster VI) but does not have a major effect on the other cluster
I mAbs tested or those from other epitope clusters.
In summary, different mAbs can lose their neutralization activ-
ity when confronted with different spike mutations, and the ef-
fects are not delineated strictly by binding clusters, so mAbs in
the same competition cluster are frequently affected differen-
tially. The triple substitution TEI470-2NVP has the most detri-
mental effect on different antibodies and affects mAbs from
nearly all binding clusters, and S494D also affects many different
clusters.
Effect of spike mutations on serum neutralization
Following identification of seven spike mutations that can limit or
abrogate neutralizing activity of mAbs (Figure 2A), the next step
was to assess the effect of these mutations on serum neutraliza-
tion. Samples were tested following two different scenarios: from
a previously characterized cohort of healthcare workers who
experienced mild COVID-19 (Houlihan et al., 2020) and sera
from a cohort of hospitalized individuals who experienced severe
COVID-19. Samples from both cohorts were collected between
March and July 2020. Eighteen samples were chosen from
both cohorts to obtain representatives with intermediate
(1:100–1,000) and potent (>1:1,000) neutralizing 50% inhibitory
dilution (ID50) values (Figure 2B). Strikingly, serum samples from
both cohorts are less affected by spike mutations than individual
mAbs in terms of fold decrease in neutralization potency (Figures
2C and 2D). Only one of 36 serum samples lost all neutralizing ac-
tivity, in contrast to the five mAbs from five different epitope clus-
ters where neutralization was abrogated completely by a single
spike mutation (Figure 1C). Moreover, the fold decrease in
neutralization potency was more modest for sera than mAbs,
with an average 3-fold decrease across all sera for the most
disadvantageous mutation, TEI470-2NVP, compared with a
more than 100-fold decrease observed for several of the mAbs
(Figures 2C and 2D). Interestingly, only one of the 36 serum sam-
ples lost more than 3-fold potency against the other triple substi-
4 Cell Reports 34, 108890, March 23, 2021
Report
tution, KVG444-6TST, which is consistent with recent data
showing that a single mutation at G446 caused a major loss of
neutralization in one sample (Greaney et al., 2021). Importantly,
there was a notable difference between the resilience of serum
samples from severely ill, hospitalized individuals and those
who experienced mild illness. Only three serum samples from
hospitalized individuals lost more than 3-fold potency against
any individual mutant (Figure 2D), whereas half of the mild illness
serum samples showed a 3-fold drop in potency against at least
one spike mutant (Figure 2D).
Greater levels of spike-reactive antibodies in sera after
severe illness
The differences in resilience to spike mutations seen in the
neutralizing sera from these two infection scenarios is plausibly
due to greater polyclonality arising from greater antigenic stimu-
lation during severe illness. To assess the serological profiles of
these two cohorts, we compared the ID50 values across 192 sam-
ples and measured the binding titers by semiquantitative ELISA
for 199 samples, as described previously (Ng et al., 2020;
O’Nions et al., 2021). There are significantly higher median immu-
noglobulin G (IgG) binding titers and median ID50 in the severe
illness cohort compared with the mild illness group (Figures 3A
and 3B; Figure S2), in line with previous observations (Seow
et al., 2020). However, when considering how the IgG binding titer
from each individual relates to the neutralization titer, it became
clear that there was a discrepancy (Figures 3C and 3D). Most hos-
pitalized individuals required a binding titer of more than 10 mg/
mL to achieve strong neutralization (ID50 > 100). Moreover, mild
infection could lead to potent neutralization (ID50 > 1,000) at bind-
ing titers of less than 10 mg/mL (Figure 3D), whereas this was
observed for only two individuals following severe illness (Fig-
ure 3C). In fact, the amount of specific IgG present at the serum
ID50 is significantly higher with severe illness compared with
mild disease (Figure 3E).
Effect of spike variants on mAb and serum
neutralization
Investigating the ability of post-SARS-CoV-2 infection mAbs and
serum to cope with mutations based on differences with SARS-
CoV was a rational first approach to study escape because these
mutations were likely to form viable spike proteins. However, to
date, none of the mutations engineered in our study have been
observed more than 20 times in global SARS-CoV-2 sequences,
although other amino acid substitutions have occurred at these
positions, including one change (L452R) that has been observed
more than 1,000 times. However, additional viral variants have
started to emerge on a significant scale (Li et al., 2020; Weisblum
et al., 2020), such as the D614G mutation, observed in western
Europe in February 2020 and now dominant across the globe
(Korber et al., 2020). More recently, a new variant, B.1.1.7, has
emerged in England and is associated with a rapid rise in case
numbers (Kemp et al., 2020; Rambaut et al., 2020). B.1.1.7 en-
codes nine sites of change in spike relative to the original Wuhan
strain. Of these, the most likely candidates to alter neutralization
sensitivity are the deletion in the NTD (DH69/V70) and the N501Y
substitution in the RBM (Kemp et al., 2020; Rambaut et al.,
2020). Therefore, we introduced these changes into the
 ll
Report OPEN ACCESS

Figure 2. Neutralization by serum is affected less adversely by SARS-CoV amino acid substitutions in SARS-CoV-2 spike
(A) Representation of the SARS CoV-2 spike trimer (blue) in complex with ACE-2 (pink) (PDB: 7DF4). The magnified image shows mutated amino acid side chains
at residues of interest.
(B) Thirty-six serum samples were assessed by pseudotyped neutralization assay. Average ID50 values for 3 independent repeats are linked by horizontal bars for
each individual sample.
(C) Fold decrease in IC50 values for each mAb against each mutant pseudotype relative to the SARS-CoV-2 wild-type pseudotype. Competitive binding clusters of
each mAb that loses more than 3-fold neutralization activity are labeled.
(D) The y axis shows the fold decrease in ID50 values for each serum sample against each mutant pseudotype relative to the SARS-CoV-2 wild-type pseudotype;
the group of affected individuals is indicated above each graph.
(C and D) The dotted horizontal lines indicate a 3-fold drop in neutralization potency.

Wuhan-strain spike in the presence of D614G. We found that
DH69/V70 did not affect the neutralization potency of most of
the mAbs tested, including COVA2-17 (Figure 4A), which binds
the RBD and NTD (Rosa et al., 2021). The exception was the
structurally unmapped COVA1-21, as reported previously
(Kemp et al., 2021). Similarly, no major drop in serum neutraliza-
tion was observed against DH69/V70 (Figure 4B). In contrast,
introduction of the N501Y substitution dramatically lowered the
neutralization potency of COVA1-12 with a fold decrease in
IC50 of more than 40 (Figures 4A and 4C). Moreover, a 5-fold

Cell Reports 34, 108890, March 23, 2021 5

 ll
OPEN ACCESS
decrease in COVA2-17 potency was observed against the N501Y
pseudotype. However, as seen for the other mutations that abro-
gate mAb function, the N501Y change had less of an effect on
sera obtained after severe and mild infection (Figures 4B and
4C).
Effect of B.1.1.7 spike on mAb and serum neutralization
Finally, a B.1.1.7 spike pseudotyping plasmid was synthesized
to incorporate the mutations observed in this new variant
(DH69/V70, DY144, N501Y, A570D, D614G, P681H, T716I, S982A, and
D1118H). This showed that, similar to the individual N501Y and
DH69/V70 mutants, B.1.1.7 can lessen the potency of three
mAbs: COVA2-17, COVA1-12, and COVA1-21 (Figure 4A). These
belong to distinct clusters and so do not compete for binding to
the same epitope. First, COVA2-17 showed an approximate 5-
fold drop in potency against the N501Y single mutant and the
B.1.1.7 pseudotype, implying that this loss of potency is primarily
N501Y driven. In contrast, the decrease in COVA1-12 potency
noted with the single N501Y change was less profound against
B.1.1.7. Furthermore, COVA1-21 experienced a substantial
drop in potency against B.1.1.7. This mAb, which does not
bind to RBD or S1 subunits, lost potency by more than 100-
fold. The B.1.1.7 pseudotype was then tested against the 36
serum samples (Figures 4B and 4C). The maximum fold
decrease in potency for the serum samples from mild illness
was 10, but the majority of samples showed less than a 3-fold
6 Cell Reports 34, 108890, March 23, 2021
Report
Figure 3. Serum responses following se-
vere COVID-19 have greater polyclonality
but less efficient neutralization
(A) The y axis shows spike S1 subunit semi-
quantitative titers measured by ELISA (STAR
Methods).
(B) The y axis shows ID50 values measured by
pseudotyped neutralization assay.
(C and D) ID50 values for serum samples versus the
corresponding S1 IgG binding titer. The relative
ranking of neutralization titers is indicated in the
graph.
(E) Concentrations of S1-specific serum IgG (pi-
cograms) at ID50 dilutions were calculated using
the IgG titers quantified via semiquantitative
ELISA.
Data for (A), (B), and (E) were analyzed by a non-
parametric Mann-Whitney U test; ****p < 0.05.
Data were measured in duplicate. Mild and severe
illness groups are defined in STAR Methods. See
also Figure S2.
change. Similarly, the maximum decrease seen for samples
from hospitalized individuals was 10-fold, but most of the sam-
ples showed a minimal change in neutralization potency. Ten
samples (28%) showed a 3- to 10-fold reduction, but because
they were potently neutralizing sera, the reduced ID50 values
were still more than 1:100 with an average reduced ID50 of
1:523, with only two samples having an ID50 of less than 1:200.
DISCUSSION
This study demonstrates that spike mutations can diminish or
abolish neutralizing activity by individual mAbs, but that serum
neutralization is affected less strongly. Notably, no serum sam-
ple failed to neutralize B.1.1.7, and only one engineered mutation
resulted in complete escape from neutralizing activity from just
one serum sample. The spike mutants evaluated comprise seven
substitutions designed to mimic possible escape changes based
on homology with SARS-CoV, two observed high-frequency mu-
tations, and the B.1.1.7 spike variant. The observation of a
modest reduction in neutralization potency against B.1.1.7 by
convalescent sera is consistent with concurrent reports (Collier
et al., 2021; Hu et al., 2021; Shen et al., 2021). The most likely
explanation for the greater effect on mAbs compared with sera
is the inherent polyclonality underlying serum neutralization.
This concept is supported by the observation that single spike
mutations can weaken neutralization for a particular mAb but
 ll
Report OPEN ACCESS

Figure 4. Variant B.1.1.7 SARS-CoV-2 spike effect on mAb and serum neutralization
(A) The indicated mAbs were assessed by pseudotype neutralization assay. Data are representative of three independent repeats. The horizontal dotted line in
each graph indicates 50% neutralization.
(B) Thirty-six serum samples (mild illness, left; severe illness, right) were assessed by pseudotype neutralization assay. ID50 values are linked by horizontal bars for
each individual sample.
(C) Fold decrease in average ID50 values from 3 repeats for each serum sample against each mutant pseudotype versus D614G. The dotted horizontal line
indicates a 3-fold drop in neutralization potency.

not for other mAbs in the same binding cluster. This highlights
that different antibodies use distinct molecular contacts within
shared epitopes so that a single mutation may not be detrimental
to all antibodies in the same binding cluster. Thus, because poly-
clonal sera contain multiple antibodies that target the major
neutralizing sites in subtly different ways, they are less sensitive
to spike mutations.
The spike mutations studied here were designed to identify po-
tential escape variants by mimicking in part the natural variation
observed between SARS-CoV and SARS-CoV-2 and are focused
mainly on the RBD as the major site of neutralizing antibody activ-
ity. Therefore, it was not surprising that many of the RBD-specific
mAbs evaluated here lost neutralization activity against one or
more of these mutations. For example, COVA2-07 and COVA2-
04 lose potency against the K417V pseudotyped virus. COVA2-
04 belongs to the VH3-53 ‘‘public’’ B cell receptor against
SARS-CoV-2 identified from multiple human infections. Thus,
COVA2-04-like antibodies are thought to be widespread among
the seropositive population, but despite this, serum samples
from mild infection showed very little change in neutralization po-
tency with K417V pseudotyped virus. Interestingly, the strongest
effect on serum samples from mild infection was mediated by
the TEI470-2NVP substitution, which is part of what has been
termed the RBD binding ridge (Greaney et al., 2021). Any muta-
tion in this zone should be monitored closely in virus populations
because of the potential for escape. Notably, the mutations that
most substantially decrease serum neutralization are those that
negatively affect mAb activity against the widest range of clusters
(I, III, XI, IX, and VI), suggesting that mAb screening is a useful
proxy for potential serum effects when a range of antibody clones
is used. However, the capacity to predict the in vivo effect of a
drop in neutralization potency requires correlation of in vitro
serum neutralization ID50 values with protection, which so far
has only been achieved in animal models where, encouragingly,
an ID50 value of 1:50 was found to be protective (McMahan et al.,
2021).

Cell Reports 34, 108890, March 23, 2021 7

 ll
OPEN ACCESS
A caveat of the first part of this study is that only RBD substi-
tutions were considered. Further studies to assess potential mu-
tations before they arise should include those in the NTD, given
the emerging importance of the NTD as a site for neutralizing an-
tibodies (Andreano et al., 2020; Rosa et al., 2021). A further lim-
itation of our original approach is that the exact mutations eval-
uated have not yet been found to any great degree in circulating
virus populations. To understand whether the conclusions from
studying the effect of the SARS-CoV-2/SARS-CoV substitutions
on neutralization parallel those of real-world spike mutations, we
examined the responses to the newly emerged B.1.1.7 variant
(Kemp et al., 2020; Rambaut et al., 2020). The RBD mutation
N501Y, shared between B.1.1.7, B.1.351, and P.1, did remove
almost all neutralizing activity for one mAb, but, in a pattern
similar to other substitutions, this did not translate into any large
effect on serum potency. We did not study the changes at posi-
tion 484 that have been observed in B.1.351 and P.1. Recently,
pseudotyped and live B.1.351 have been shown to be resistant
to neutralization by a large proportion of convalescent plasma
samples (Cele et al., 2021; Wibmer et al., 2021).
Theoretically, it is likely that combinations of mutations have
more potential to lead to loss of serum activity than single amino
acid changes by destroying multiple parts of key epitopes. This
has been observed partially in terms of the B.1.1.7 spike pseudo-
type analyzed here. Only one mAb was affected more dramati-
cally by the full set of B.1.1.7 mutations compared with the
DH69/V70 and N501Y individual mutations. However, serum sam-
ples with reduced neutralization were affected more strongly by
B.1.1.7 (Figures 4B and 4C). Importantly, these samples were
collected prior to July 2020 and therefore are highly unlikely to
be derived from B.1.1.7 infection. However, all of the affected
samples were still able to neutralize B.1.1.7, and the average
reduced serum ID50 value was 1:523. This is 10 times higher
than the reported serum ID50 correlate of protection in animal
studies (McMahan et al., 2021) and suggests that these re-
sponses would likely still be effective against infection with
B.1.1.7.
This study underlines the potential for escape from neutralizing
antibodies because of mutations in spike and the relative resil-
ience of serum responses compared with individual mAbs.
This difference likely derives from the breadth inherent in poly-
clonal sera compared with the precision interaction of a given
mAb. Our results suggest that the majority of vaccine responses
should be effective against the B.1.1.7 variant because the sera
evaluated were obtained after infection early in 2020, when the
commonly circulating virus was highly similar in sequence to
the vaccines now being deployed. These findings are in agree-
ment with concurrent studies that have reported a minimal
drop in neutralization potency against B.1.1.7 in vaccinee and/
or convalescent sera (Muik et al., 2021; Shen et al., 2021;
Wang et al., 2021b; Wu et al., 2021). Finally, because SARS-
CoV-2 seropositivity will increase across the human population
(because of vaccination efforts and natural infection), there
may be selection for spike mutations that result in substantial
antigenic drift. Recent data showing limited serum neutralization
against B.1.351 (Cele et al., 2021; Wibmer et al., 2021) suggest
that major antigenic drift has already occurred. Vaccine-induced
responses appear to be more resilient to the mutations in
8 Cell Reports 34, 108890, March 23, 2021
Report
B.1.351, in part because of higher initial titers (Collier et al.,
2021; Wang et al., 2021a, 2021b; Wu et al., 2021). However,
that this level of antigenic change has already occurred in
SARS-CoV-2 suggests that, with increasing seroprevalence,
additional potential neutralization escape mutations will emerge
and require scrutiny. The data here suggest that evaluation of
neutralizing mAbs from non-overlapping binding clusters can
highlight which spike mutations will most affect serum neutrali-
zation. Our findings stress the importance of continuous moni-
toring of variants and in vitro assessment of their effect on
neutralization. This is particularly relevant for use of convales-
cent plasma and development of therapeutic mAbs as well as
vaccine development and implementation.
CONSORTIA
The SAFER Study Investigators are Sajida Adam, Matthew By-
ott, Tom Byrne, Elise Crayton, Claudia Davies, Sarah Edwards,
Louise Enfield, Daniel Frampton, Kathleen Gärtner, Richard Gil-
son, Triantafylia Gkouleli, Nick Gotts, Andrew Hayward, Judith
Heaney, Catherine F. Houlihan, Dan Lewer, Fabiana Lorencatto,
Hinal Lukha, Ed Manley, Rebecca Matthews, Hazel McBain, An-
gela McBride, Laura E. McCoy, Carly Meyer, Susan Michie, Eleni
Nastouli, Stavroula M Paraskevopoulou, Paulina Prymas,
Veronica Ranieri, Emilie Sanchez, Abigail Severn, Maryam Shah-
manesh, Robert Shortman, Moira J. Spyer, Andrea Stoltenberg,
Nina Vora, Naomi Walker, Bethany Williams, Jared Wilson-Ag-
garwal, and Leigh Wood.
STAR+METHODS
Detailed methods are provided in the online version of this paper
and include the following:
d KEY RESOURCES TABLE
d RESOURCE AVAILABILITY
B Lead contact
B Materials availability
B Data and code availability
d EXPERIMENTAL MODEL AND SUBJECT DETAILS
B Mild illness serum samples
B Severe illness serum samples
B Bacterial Strains and Cell Culture
d METHOD DETAILS
B Spike mutant generation
B Neutralization assay
B Semiquantitative ELISA
d QUANTIFICATION AND STATISTICAL ANALYSIS
SUPPLEMENTAL INFORMATION
Supplemental Information can be found online at https://doi.org/10.1016/j.
celrep.2021.108890.
ACKNOWLEDGMENTS
The authors would like to thank James E. Voss for the gift of HeLa ACE2-ex-
pressing cells and George Kassiotis, Dan Frampton, Ann-Kathrin Reuschl,
and Joe Grove for critical feedback. We are indebted to the Biobank staff

Report
and study participants and their families at the Royal Free Hospital and the
UCLH SAFER study recruitment team and study participants. L.E.M. is sup-
ported by a Medical Research Council career development award (MR/
R008698/1). M.J.v.G. is a recipient of an AMC fellowship, and R.W.S. is a
recipient of a Vici grant from the Netherlands Organization for Scientific
Research (NWO). C.G. is supported by the MRC-KCL Doctoral Training Part-
nership in Biomedical Sciences (MR/N013700/1). This study was also funded
by the UCL Coronavirus Response Fund, made possible through generous do-
nations from UCL’s supporters, alumni, and friends (to L.E.M.); the King’s
Together Rapid COVID-19 Call award (to K.J.D.); the Huo Family Foundation
(to K.J.D.); the Royal Free Charity; and the UK Coronavirus Immunology Con-
sortium. This work was also supported by National Institutes of Health grant
P01 AI110657 and Bill and Melinda Gates Foundation grant INV-002022 (to
R.W.S.). The work in laboratory of P.C. was supported by the Francis Crick
Institute (FC001061), which receives its core funding from Cancer Research
UK, the UK Medical Research Council, and the Wellcome Trust. The SAFER
study was funded by MRC UKRI (MC_PC_19082) and supported by the
UCLH/UCL NIHR BRC.
AUTHOR CONTRIBUTIONS
K.J.D., L.E.M., L.M., S.A.G., P.T., N.L., and C.R.-S. characterized monoclonal
antibodies and sera. A.R., C.R., and P.C. expressed and purified proteins.
J.H., C.F.H., H.J.S., and E.N. assembled the panels of human serum samples.
M.J.v.G., R.W.S., Y.A., and J.L.S. isolated and provided monoclonal anti-
bodies. R.K.G. generated and provided mutated spike plasmids. M.J.v.G.,
K.J.D., and L.E.M. wrote the paper with contributions from all authors.
DECLARATION OF INTERESTS
Amsterdam UMC submitted a patent application on SARS-CoV-2 monoclonal
antibodies, some of which were used in this study.
Received: January 15, 2021
Revised: February 17, 2021
Accepted: March 1, 2021
Published: March 6, 2021
REFERENCES
Andreano, E., Piccini, G., Licastro, D., Casalino, L., Johnson, N.V., Paciello, I.,
Monego, S.D., Pantano, E., Manganaro, N., Manenti, A., et al. (2020). SARS-
CoV-2 escape in vitro from a highly neutralizing COVID-19 convalescent
plasma. bioRxiv. https://doi.org/10.1101/2020.12.28.424451.
Brouwer, P.J.M., Caniels, T.G., van der Straten, K., Snitselaar, J.L., Aldon, Y.,
Bangaru, S., Torres, J.L., Okba, N.M.A., Claireaux, M., Kerster, G., et al. (2020).
Potent neutralizing antibodies from COVID-19 patients define multiple targets
of vulnerability. Science 369, 643–650.
Burton, D.R., Poignard, P., Stanfield, R.L., and Wilson, I.A. (2012). Broadly
neutralizing antibodies present new prospects to counter highly antigenically
diverse viruses. Science 337, 183–186.
Cao, Y., Su, B., Guo, X., Sun, W., Deng, Y., Bao, L., Zhu, Q., Zhang, X., Zheng,
Y., Geng, C., et al. (2020). Potent Neutralizing Antibodies against SARS-CoV-2
Identified by High-Throughput Single-Cell Sequencing of Convalescent Pa-
tients’ B Cells. Cell 182, 73–84.e16.
Cele, S., Gazy, I., Jackson, L., Hwa, S.-H., Tegally, H., Lustig, G., Giandhari, J.,
Pillay, S., Wilkinson, E., Naidoo, Y., et al. (2021). Escape of SARS-CoV-2
501Y.V2 variants from neutralization by convalescent plasma. medRxiv.
https://doi.org/10.1101/2021.01.26.21250224.
Centers for Disease Control and Prevention (CDC) (2003). Update: Outbreak of
severe acute respiratory syndrome–worldwide, 2003. MMWR Morb. Mortal.
Wkly. Rep. 52, 241–246, 248.
Cevik, M., Kuppalli, K., Kindrachuk, J., and Peiris, M. (2020). Virology, trans-
mission, and pathogenesis of SARS-CoV-2. BMJ 371, m3862.
ll
OPEN ACCESS
Collier, D.A., De Marco, A., Ferreira, I.A.T.M., Meng, B., Datir, R., Walls, A.C.,
Kemp, S.A., Bassi, J., Pinto, D., Fregni, C.S., et al. (2021). SARS-CoV-2 B.1.1.7
escape from mRNA vaccine-elicited neutralizing antibodies. medRxiv. https://
doi.org/10.1101/2021.01.19.21249840.
Faria, N.R., Claro, I.M., Candido, D., Moyses Franco, L.A., Andrade, P.S., Co-
letti, T.M., Silva, C.A.M., Sales, F.C., Manuli, E.R., Aguiar, R.S., et al. (2021).
Genomic characterisation of an emergent SARS-CoV-2 lineage in Manaus:
preliminary findings. https://virological.org/t/genomic-characterisation-of-an-
emergent-sars-cov-2-lineage-in-manaus-preliminary-findings/586.
Gralinski, L.E., and Menachery, V.D. (2020). Return of the Coronavirus: 2019-
nCoV. Viruses 12, 135.
Greaney, A.J., Loes, A.N., Crawford, K.H.D., Starr, T.N., Malone, K.D., Chu,
H.Y., and Bloom, J.D. (2021). Comprehensive mapping of mutations to the
SARS-CoV-2 receptor-binding domain that affect recognition by polyclonal
human serum antibodies. bioRxiv. https://doi.org/10.1101/2020.12.31.
425021.
Houlihan, C.F., Vora, N., Byrne, T., Lewer, D., Kelly, G., Heaney, J., Gandhi, S.,
Spyer, M.J., Beale, R., Cherepanov, P., et al.; Crick COVID-19 Consortium;
SAFER Investigators (2020). Pandemic peak SARS-CoV-2 infection and sero-
conversion rates in London frontline health-care workers. Lancet 396, e6–e7.
Hu, J., Peng, P., Wang, K., Liu, B.-z., Fang, L., Luo, F.-y., Jin, A.-s., Tang, N.,
and Huang, A.-l. (2021). Emerging SARS-CoV-2 variants reduce neutralization
sensitivity to convalescent sera and monoclonal antibodies. bioRxiv. https://
doi.org/10.1101/2021.01.22.427749.
Kemp, S., Datir, R., Collier, D., Ferreira, I., Carabelli, A., Harvey, W., Robertson,
D., and Gupta, R. (2020). Recurrent emergence and transmission of a SARS-
CoV-2 Spike deletion DH69/DV70. bioRxiv. https://doi.org/10.1101/2020.12.
14.422555.
Kemp, S.A., Collier, D.A., Datir, R.P., Ferreira, I.A.T.M., Gayed, S., Jahun, A.,
Hosmillo, M., Rees-Spear, C., Mlcochova, P., Lumb, I.U., et al.; CITIID-NIHR
BioResource COVID-19 Collaboration; COVID-19 Genomics UK (COG-UK)
Consortium (2021). SARS-CoV-2 evolution during treatment of chronic infec-
tion. Nature, Published online February 5, 2021. https://doi.org/10.1038/
s41586-021-03291-y.
Korber, B., Fischer, W.M., Gnanakaran, S., Yoon, H., Theiler, J., Abfalterer, W.,
Hengartner, N., Giorgi, E.E., Bhattacharya, T., Foley, B., et al.; Sheffield
COVID-19 Genomics Group (2020). Tracking Changes in SARS-CoV-2 Spike:
Evidence that D614G Increases Infectivity of the COVID-19 Virus. Cell 182,
812–827.e19.
Lan, J., Ge, J., Yu, J., Shan, S., Zhou, H., Fan, S., Zhang, Q., Shi, X., Wang, Q.,
Zhang, L., and Wang, X. (2020). Structure of the SARS-CoV-2 spike receptor-
binding domain bound to the ACE2 receptor. Nature 581, 215–220.
Li, W., Moore, M.J., Vasilieva, N., Sui, J., Wong, S.K., Berne, M.A., Somasun-
daran, M., Sullivan, J.L., Luzuriaga, K., Greenough, T.C., et al. (2003). Angio-
tensin-converting enzyme 2 is a functional receptor for the SARS coronavirus.
Nature 426, 450–454.
Li, Q., Wu, J., Nie, J., Zhang, L., Hao, H., Liu, S., Zhao, C., Zhang, Q., Liu, H.,
Nie, L., et al. (2020). The Impact of Mutations in SARS-CoV-2 Spike on Viral
Infectivity and Antigenicity. Cell 182, 1284–1294.e9.
Lipsitch, M., Cohen, T., Cooper, B., Robins, J.M., Ma, S., James, L., Gopalak-
rishna, G., Chew, S.K., Tan, C.C., Samore, M.H., et al. (2003). Transmission dy-
namics and control of severe acute respiratory syndrome. Science 300, 1966–
1970.
Liu, H., Wu, N.C., Yuan, M., Bangaru, S., Torres, J.L., Caniels, T.G., van Schoo-
ten, J., Zhu, X., Lee, C.D., Brouwer, P.J.M., et al. (2020). Cross-Neutralization
of a SARS-CoV-2 Antibody to a Functionally Conserved Site Is Mediated by
Avidity. Immunity 53, 1272–1280.e5.
McMahan, K., Yu, J., Mercado, N.B., Loos, C., Tostanoski, L.H., Chandrashe-
kar, A., Liu, J., Peter, L., Atyeo, C., Zhu, A., et al. (2021). Correlates of protec-
tion against SARS-CoV-2 in rhesus macaques. Nature 590, 630–634.
Mor, M., Werbner, M., Alter, J., Safra, M., Chomsky, E., Hada-Neeman, S., Po-
lonsky, K., Nowell, C.J., Clark, A.E., Roitburd-Berman, A., et al. (2020). Multi-
Clonal Live SARS-CoV-2 In Vitro Neutralization by Antibodies Isolated from
Cell Reports 34, 108890, March 23, 2021 9
 ll
OPEN ACCESS
Severe COVID-19 Convalescent Donors. bioRxiv. https://doi.org/10.1101/
2020.10.06.323634.
Muik, A., Wallisch, A.-K., Sänger, B., Swanson, K.A., Mu €hl, J., Chen, W., Cai,
H., Maurus, D., Sarkar, R., Tu €reci, Ö., et al. (2021). Neutralization of SARS-
CoV-2 lineage B.1.1.7 pseudovirus by BNT162b2 vaccine-elicited human
sera. Science 29, eabg6105.
Ng, K.W., Faulkner, N., Cornish, G.H., Rosa, A., Harvey, R., Hussain, S., Ul-
ferts, R., Earl, C., Wrobel, A.G., Benton, D.J., et al. (2020). Preexisting and
de novo humoral immunity to SARS-CoV-2 in humans. Science 370, 1339–
1343.
O’Nions, J., Muir, L., Zheng, J., Rees-Spear, C., Rosa, A., Roustan, C., Earl, C.,
Cherepanov, P., Gupta, R., Khwaja, A., et al. (2021). SARS-CoV-2 antibody re-
sponses in patients with acute leukaemia. Leukemia 35, 289–292.
Ortega, J.T., Serrano, M.L., Pujol, F.H., and Rangel, H.R. (2020). Role of
changes in SARS-CoV-2 spike protein in the interaction with the human
ACE2 receptor: An in silico analysis. EXCLI J. 19, 410–417.
Oude Munnink, B.B., Sikkema, R.S., Nieuwenhuijse, D.F., Molenaar, R.J.,
Munger, E., Molenkamp, R., van der Spek, A., Tolsma, P., Rietveld, A.,
Brouwer, M., et al. (2021). Transmission of SARS-CoV-2 on mink farms be-
tween humans and mink and back to humans. Science 371, 172–177.
Petersen, E., Koopmans, M., Go, U., Hamer, D.H., Petrosillo, N., Castelli, F.,
Storgaard, M., Al Khalili, S., and Simonsen, L. (2020). Comparing SARS-
CoV-2 with SARS-CoV and influenza pandemics. Lancet Infect. Dis. 20,
e238–e244.
Pettersen, E.F., Goddard, T.D., Huang, C.C., Couch, G.S., Greenblatt, D.M.,
Meng, E.C., and Ferrin, T.E. (2004). UCSF Chimera–a visualization system
for exploratory research and analysis. J. Comput. Chem. 25, 1605–1612.
Plotkin, S.A. (2008). Vaccines: correlates of vaccine-induced immunity. Clin.
Infect. Dis. 47, 401–409.
Polack, F.P., Thomas, S.J., Kitchin, N., Absalon, J., Gurtman, A., Lockhart, S.,
Perez, J.L., Pérez Marc, G., Moreira, E.D., Zerbini, C., et al.; C4591001 Clinical
Trial Group (2020). Safety and Efficacy of the BNT162b2 mRNA Covid-19 Vac-
cine. N. Engl. J. Med. 383, 2603–2615.
Rambaut, A., Loman, N., Pybus, O., Barclay, W., Barrett, J., Carabelli, A., Con-
nor, T., Peacock, T., Robertson, D.L., and Volz, E. (2020). Preliminary genomic
characterisation of an emergent SARS-CoV-2 lineage in the UK defined by a
novel set of spike mutations. https://virological.org/t/preliminary-genomic-
characterisation-of-an-emergent-sars-cov-2-lineage-in-the-uk-defined-by-a-
novel-set-of-spike-mutations/563.
Robbiani, D.F., Gaebler, C., Muecksch, F., Lorenzi, J.C.C., Wang, Z., Cho, A.,
Agudelo, M., Barnes, C.O., Gazumyan, A., Finkin, S., et al. (2020). Convergent
antibody responses to SARS-CoV-2 in convalescent individuals. Nature 584,
437–442.
Rogers, T.F., Zhao, F., Huang, D., Beutler, N., Burns, A., He, W.T., Limbo, O.,
Smith, C., Song, G., Woehl, J., et al. (2020). Isolation of potent SARS-CoV-2
neutralizing antibodies and protection from disease in a small animal model.
Science 369, 956–963.
Rosa, A., Pye, V.E., Graham, C., Muir, L., Seow, J., Ng, K.W., Cook, N.J., Rees-
Spear, C., Parker, E., dos Santos, M.S., et al. (2021). SARS-CoV-2 recruits a
haem metabolite to evade antibody immunity. medRxiv. https://doi.org/10.
1101/2021.01.21.21249203.
Seow, J., Graham, C., Merrick, B., Acors, S., Pickering, S., Steel, K.J.A., Hem-
mings, O., O’Byrne, A., Kouphou, N., Galao, R.P., et al. (2020). Longitudinal
observation and decline of neutralizing antibody responses in the three months
following SARS-CoV-2 infection in humans. Nat. Microbiol. 5, 1598–1607.
Shen, X., Tang, H., McDanal, C., Wagh, K., Fischer, W., Theiler, J., Yoon, H., Li,
D., Haynes, B.F., Sanders, K.O., et al. (2021). SARS-CoV-2 variant B.1.1.7 is
susceptible to neutralizing antibodies elicited by ancestral Spike vaccines. bio-
Rxiv. https://doi.org/10.1101/2021.01.27.428516.
10 Cell Reports 34, 108890, March 23, 2021
Report
Tegally, H., Wilkinson, E., Giovanetti, M., Iranzadeh, A., Fonseca, V., Giand-
hari, J., Doolabh, D., Pillay, S., San, E.J., Msomi, N., et al. (2020). Emergence
and rapid spread of a new severe acute respiratory syndrome-related corona-
virus 2 (SARS-CoV-2) lineage with multiple spike mutations in South Africa.
medRxiv. https://doi.org/10.1101/2020.12.21.20248640.
Varabyou, A., Pockrandt, C., Salzberg, S.L., and Pertea, M. (2020). Rapid
detection of inter-clade recombination in SARS-CoV-2 with Bolotie. bioRxiv.
https://doi.org/10.1101/2020.09.21.300913.
Wang, P., Nair, M.S., Liu, L., Iketani, S., Luo, Y., Guo, Y., Wang, M., Yu, J.,
Zhang, B., Kwong, P.D., et al. (2021a). Antibody Resistance of SARS-CoV-2
Variants B.1.351 and B.1.1.7. bioRxiv. https://doi.org/10.1101/2021.01.25.
428137.
Wang, Z., Schmidt, F., Weisblum, Y., Muecksch, F., Barnes, C.O., Finkin, S.,
Schaefer-Babajew, D., Cipolla, M., Gaebler, C., Lieberman, J.A., et al.
(2021b). mRNA vaccine-elicited antibodies to SARS-CoV-2 and circulating
variants. bioRxiv. https://doi.org/10.1038/s41586-021-03324-6.
Weisblum, Y., Schmidt, F., Zhang, F., DaSilva, J., Poston, D., Lorenzi, J.C.,
Muecksch, F., Rutkowska, M., Hoffmann, H.H., Michailidis, E., et al. (2020).
Escape from neutralizing antibodies by SARS-CoV-2 spike protein variants.
eLife 9, e61312.
Welkers, M.R.A., Han, A.X., Reusken, C.B.E.M., and Eggink, D. (2021).
Possible host-adaptation of SARS-CoV-2 due to improved ACE2 receptor
binding in mink. Virus Evol 7, veaa094.
Wibmer, C.K., Ayres, F., Hermanus, T., Madzivhandila, M., Kgagudi, P., Lamb-
son, B.E., Vermeulen, M., van den Berg, K., Rossouw, T., Boswell, M., et al.
(2021). SARS-CoV-2 501Y.V2 escapes neutralization by South African
COVID-19 donor plasma. bioRxiv. https://doi.org/10.1101/2021.01.18.
427166.
Wu, N.C., Yuan, M., Bangaru, S., Huang, D., Zhu, X., Lee, C.D., Turner, H.L.,
Peng, L., Yang, L., Burton, D.R., et al. (2020). A natural mutation between
SARS-CoV-2 and SARS-CoV determines neutralization by a cross-reactive
antibody. PLoS Pathog. 16, e1009089.
Wu, K., Werner, A.P., Moliva, J.I., Koch, M., Choi, A., Stewart-Jones, G.B.E.,
Bennett, H., Boyoglu-Barnum, S., Shi, W., Graham, B.S., et al. (2021).
mRNA-1273 vaccine induces neutralizing antibodies against spike mutants
from global SARS-CoV-2 variants. bioRxiv. https://doi.org/10.1101/2021.01.
25.427948.
Xie, X., Liu, Y., Liu, J., Zhang, X., Zou, J., Fontes-Garfias, C.R., Xia, H., Swan-
son, K.A., Cutler, M., Cooper, D., et al. (2021). Neutralization of SARS-CoV-2
spike 69/70 deletion, E484K and N501Y variants by BNT162b2 vaccine-eli-
cited sera. Nat. Med. Published online February 8, 2021. https://doi.org/10.
1038/s41591-021-01270-4.
Xu, C., Wang, Y., Liu, C., Zhang, C., Han, W., Hong, X., Wang, Y., Hong, Q.,
Wang, S., Zhao, Q., et al. (2021). Conformational dynamics of SARS-CoV-2
trimeric spike glycoprotein in complex with receptor ACE2 revealed by cryo-
EM. Sci. Adv. 7, eabe5575.
Yi, C., Sun, X., Ye, J., Ding, L., Liu, M., Yang, Z., Lu, X., Zhang, Y., Ma, L., Gu,
W., et al. (2020). Key residues of the receptor binding motif in the spike protein
of SARS-CoV-2 that interact with ACE2 and neutralizing antibodies. Cell. Mol.
Immunol. 17, 621–630.
Zost, S.J., Wu, N.C., Hensley, S.E., and Wilson, I.A. (2019). Immunodominance
and Antigenic Variation of Influenza Virus Hemagglutinin: Implications for
Design of Universal Vaccine Immunogens. J. Infect. Dis. 219 (Suppl_1),
S38–S45.
Zufferey, R., Nagy, D., Mandel, R.J., Naldini, L., and Trono, D. (1997). Multiply
attenuated lentiviral vector achieves efficient gene delivery in vivo. Nat. Bio-
technol. 15, 871–875.
 ll
Report OPEN ACCESS

STAR+METHODS

KEY RESOURCES TABLE

REAGENT or RESOURCE SOURCE IDENTIFIER

Antibodies
Goat anti-human F(ab)’2 Stratech Cat# 109-006-006; RRID: AB_2337553
Alkaline phosphatase-conjugated Stratech Cat#109-055-098; RRID: AB_2337608
goat anti-human IgG
Bacterial and virus strains
XL1-Blue Supercompetent Cell Agilent Cat# 210518
Biological Samples
Mild illness serum samples UCLH SAFER study; Houlihan et al., 2020 NHS Health Research Authority reference
no. 20/SC/0147
Severe illness patient serum samples Tissue Access for Patient Benefit (TAPb), Reference no. NC2020.24; NRES EC no.
The Royal Free Hospital 16/WA/0289
Chemicals, peptides, and recombinant proteins
SARS-CoV-2 spike S1 protein Peter Cherepanov Laboratory; Ng et al., N/A
2020
Critical commercial assays
Bright-Glo Luciferase kit Promega Cat# E2650
QuickChange Lightening Agilent Cat# 210518
Site-Directed Mutagenesis kit
Experimental models: cell lines
Human: HEK293T/17 cells American Type Culture Collection ATCC CRL-11268
Human: HeLa-ACE-2 cells James Voss Laboratory, The Scripps N/A
Research Institute; Rogers et al., 2020
Oligonucleotides
50 -agcaatttcagagtgcagcctaccgGg ES324-6GD Forward 1107
GAcatcgtgagattccctaatatcacc-30
50 -tacagcgtgctgtacaatagcg S373F Forward 1109
ccTTcttcagcaccttcaaatgttatggt-30
50 -accttcaaatgttatggtgtttcgGcaa P384A Forward 1111
caaagctgaatgacctgtgcttc-30
50 -cagatcgcgccagggcagaccgg K417V Forward 1112
cGTgatcgccgactacaattacaagctg-30
50 -tggaactctaacaatctagattcgaCa KVG444-6TST Forward 1114
TCtACaggcaattacaattacctgtacaga-30
50 -aaagttggaggcaattacaattacA L452K Forward 1115
Agtacagactgttcagaaagagcaat-30
50 -ggcaattacaattacctgtacagaTA LF455-6YL Forward 1116
CCtcagaaagagcaatctgaagcctttc-30
50 -aagcctttcgagagagacatcagcaAcgTg TEI470-472NVP Forward 1118
CCctaccaggccggcagcacaccgtgt-30
50 -ttcaattgctacttccctctgcag S494D Forward 1122
GActacggcttccagcctaccaatggc-30
50 -ggtgatattagggaatctcacgatg ES324-6GD Reverse 1126
TCcCcggtaggctgcactctgaaattgct-30
50 -accataacatttgaaggtgctgaagAAg S373F Reverse 1128
gcgctattgtacagcacgctgta-30
(Continued on next page)

Cell Reports 34, 108890, March 23, 2021 e1

 ll
OPEN ACCESS Report

Continued
REAGENT or RESOURCE SOURCE IDENTIFIER
0
5 -gaagcacaggtcattcagctttg P384A Reverse 1130
ttgCcgaaacaccataacatttgaaggt-30
50 -cagcttgtaattgtagtcggcga K417V Reverse 1131
tcACgccggtctgccctggcgcgatctg-30
50 -tctgtacaggtaattgtaattgcctGTa KVG444-6TST Reverse 1133
GAtGtcgaatctagattgttagagttcca-30
50 -attgctctttctgaacagtctgtacTTgta L452K Reverse 1134
attgtaattgcctccaacttt-30
50 -gaaaggcttcagattgctctttctgaGGTA LF455-6YL Reverse 1135
tctgtacaggtaattgtaattgcc-30
50 -acacggtgtgctgccggcctggta TEI470-472NVP Reverse 1137
gGGcAcgTtgctgatgtctctctcgaaaggctt-30
50 -gccattggtaggctggaagccgtag S494D Reverse 1141
TCctgcagagggaagtagcaattgaa-30
Recombinant DNA
pCDNA3.1+ D614G spike Kemp et al., 2021 pCDNA_Spike D614G
expression vector
pCDNA3.1+ D614G_DH69/V70 Spike Kemp et al., 2021 pCDNA_Spike D614G_DH69/V70
expression vector
pCDNA3.1+ D614G_N501Y spike Generated in this study pCDNA_Spike D614G_N501Y
expression vector
SARS-CoV-2 Wuhan spike Seow et al., 2020 pCDNA_Spike
pCDNA3.1+ expression vector
pCDNA3.1+ B.1.1.7 spike (DH69/V70, Synthesized by Genewiz Inc. and pCDNA_B.1.1.7
DY144, N501Y, A570D, D614G, P681H, subcloned into pcDNA3.1+
T716I, S982A, D1118H) expression vector
SARS-CoV-2 Wuhan spike pCDNA3.1+ Generated in this study P384A
expression vector P384A mutation
SARS-CoV-2 Wuhan spike pCDNA3.1+ Generated in this study K417V
expression vector K417V mutation
SARS-CoV-2 Wuhan spike pCDNA3.1+ Generated in this study KVG444-6TST
expression vector KVG446-TST mutation
SARS-CoV-2 Wuhan spike pCDNA3.1+ Generated in this study L452K
expression vector L452K mutation
SARS-CoV-2 Wuhan spike pCDNA3.1+ Generated in this study LF455-6YL
expression vector LF455-6YL mutation
SARS-CoV-2 Wuhan spike pCDNA3.1+ Generated in this study TEI470-472NVP
expression vector TEI470-2NVP mutation
SARS-CoV-2 Wuhan spike pCDNA3.1+ Generated in this study S494D
expression vector S494D mutation
SARS-CoV-2 Wuhan spike pCDNA3.1+ Generated in this study ES324-6GD
expression vector ES324-6GD mutation
SARS-CoV-2 Wuhan spike pCDNA3.1+ Generated in this study S373F
expression vector S373F mutation
HIV-1 luciferase reporter vector Seow et al., 2020 CSWL HIV-1 luciferase reporter
HIV p8.91 packaging construct Zufferey et al., 1997 p8.91
Software and Algorithms
UCSF Chimera Pettersen et al., 2004 https://www.cgl.ucsf.edu/chimera/
Prism 8 GraphPad prism https://www.graphpad.com/
scientific-software/prism/

e2 Cell Reports 34, 108890, March 23, 2021

 ll
Report OPEN ACCESS

RESOURCE AVAILABILITY

Lead contact
Further information and requests for resources and reagents should be directed to and will be fulfilled by the Lead Contact, Laura
McCoy (l.mccoy@ucl.ac.uk).

Materials availability
Reagents generated in this study are available from the Lead Contact with a completed Materials Transfer Agreement.

Data and code availability

This study did not generate datasets/code. Original source data for SARS-CoV-2 spike structure used in Figure 2 is available at
https://doi.org/10.2210/pdb7DF4/pdb, and in Figure S1 at https://doi.org/10.2210/pdb6VXX/pdb.

EXPERIMENTAL MODEL AND SUBJECT DETAILS

Mild illness serum samples

These samples are part of the UCLH SAFER study and were collected as previously described (Houlihan et al., 2020). Briefly, samples
are from 81 seropositive individuals previously identified (Houlihan et al., 2020) who donated blood at monthly intervals from March to
July 2020 as well as undergoing regular PCR testing. Informed consent was obtained from all participants. The median age of par-
ticipants was 81 (interquartile range 70-87), 43% were female and 57% male. The study protocol was approved by the NHS Health
Research Authority (ref 20/SC/0147) on 26 March 2020. Ethical oversight was provided by the South- Central Berkshire Research
Ethics Committee.

Severe illness serum samples

These samples are from patients hospitalized for COVID-19 between March and July 2020 and were obtained during their hospital
stay through the Tissue Access for Patient Benefit (TAPb) scheme at The Royal Free Hospital (approved by UCL–Royal Free Hospital
BioBank Ethical Review Committee Reference number: NC2020.24 NRES EC number: 16/WA/0289). Informed consent was ob-
tained from all participants and a single blood sample was taken without interfering with normal clinical care. The median age of par-
ticipants was 34 years (interquartile range 29–44), 62% were female and 38% male.

Bacterial Strains and Cell Culture

Bacterial transformations were performed with XL1-Blue Supercompetent Cells (Agilent). SARS-CoV-2 pseudotypes were produced by
transfection of HEK293T/17 cells and neutralization activity assayed using HeLa cells stably expressing ACE2 (Kind gift James E Voss).

METHOD DETAILS

Spike mutant generation

QuikChange Lightening Site-Directed Mutagenesis kit was used to generate amino acid substitutions in the SARS-CoV-2 Wuhan
spike expression vector (Seow et al., 2020) or the D614G pCDNA spike plasmid (Kemp et al., 2021) following the manufacturer’s in-
structions (Agilent Technologies, Inc., Santa Clara, CA). Spike B.1.1.7 (DH69/V70, DY144, N501Y, A570D, D614G, P681H, T716I, S982A,
D1118H) was synthesized by Genewiz, Inc. and cloned into the pCDNA3.1+ expression vector using BamHI and EcoRI restriction sites.

Neutralization assay
HIV-1 particles pseudotyped with SARS-CoV-2 spike were produced in a T75 flask seeded the day before with 3 million HEK293T/17
cells in 10 ml complete DMEM, supplemented with 10% FBS, 100 IU/ml penicillin and 100 mg/ml streptomycin. Cells were transfected
using 60 mg of PEI-Max (Polysciences) with a mix of three plasmids: 9.1 mg HIV-1 luciferase reporter vector (Seow et al., 2020), 9.1 mg
HIV p8.91 packaging construct and 1.4 mg WT SARS-CoV-2 spike expression vector (Seow et al., 2020). Supernatants containing
pseudotyped virions were harvested 48 h post-transfection, filtered through a 0.45-mm filter and stored at 80 C. Neutralization as-
says were conducted by serial dilution of monoclonal IgG at the indicated concentrations in DMEM (10% FBS and 1% penicillin–strep-
tomycin) and incubated with pseudotyped virus for 1 h at 37 C in 96-well plates. HeLa cells stably expressing ACE-2 (provided by J.E.
Voss, Scripps Institute) were then added to the assay (10,000 cells per 100 ml per well). After 48-72 h luminescence was assessed as a
proxy of infection by lysing cells with the Bright-Glo luciferase kit (Promega), using a Glomax plate reader (Promega). Measurements
were performed in duplicate and used to calculate 50% inhibitory dilutions/concentration (ID/C50) values in GraphPad Prism software.

Semiquantitative ELISA
As described previously (O’Nions et al., 2021) nine columns of a half-well 96-well MaxiSorp plate were coated with purified SARS-
CoV-2 spike S1 protein in PBS (3 mg/ml per well in 25 mL) and the remaining three columns were coated with 25 mL goat anti-human
F(ab)’2 diluted 1:1000 in PBS to generate the internal standard curve. After incubation at 4 C overnight, the ELISA plate was blocked

Cell Reports 34, 108890, March 23, 2021 e3

 ll
OPEN ACCESS Report

for 1 h in assay buffer (PBS, 5% milk, 0.05% Tween 20). Sera was diluted in assay buffer at dilutions from 1:50 to 1:5000 and 25 mL
added to the ELISA plate. Serial dilutions of known concentrations of IgG standards were applied to the three standard curve col-
umns in place of sera. The ELISA plate was then incubated for 2 h at room temperature and then washed 4 times with PBS-T
(PBS, 0.05% Tween 20). Alkaline phosphatase-conjugated goat anti-human IgG at a 1:1000 dilution was then added to each well
and incubated for 1 h. Following this, plates were washed 6 times with PBS-T and 25 mL of colorimetric alkaline phosphatase sub-
strate added. Absorbance was measured at 405 nm. Antigen-specific IgG concentrations in serum were then calculated based on
interpolation from the IgG standard results using a four-parameter logistic (4PL) regression curve fitting model.

QUANTIFICATION AND STATISTICAL ANALYSIS

All neutralization measurements were performed in duplicate and 50% inhibitory concentrations/dilutions (IC/ID50) were calculated
using GraphPad Prism software. ID50 values calculated as indicated in the relevant Figure legends. Statistical analysis in Figure 3
(non-parametric Mann-Whitney U test) was performed using GraphPad Prism software, significance defined as ****p < 0.05. Fold
decrease in serum ID50 was calculated by dividing the average ID50 value for a given sample against SARS-CoV-2 or SARS-CoV-
2 D614G (as indicated) by the average ID50 value for that sample against the indicated mutant or variant pseudotype. Fold decrease
in mAb IC50 was calculated by dividing the average IC50 value for a given mAb against the indicated mutant or variant pseudotype by
the average IC50 value for that mAb against the SARS-CoV-2 or SARS-CoV-2 D614G (as indicated).

e4 Cell Reports 34, 108890, March 23, 2021

Cell Reports, Volume 34

Supplemental information

The effect of spike mutations

on SARS-CoV-2 neutralization
Chloe Rees-Spear, Luke Muir, Sarah A. Griffith, Judith Heaney, Yoann Aldon, Jonne L.
Snitselaar, Peter Thomas, Carl Graham, Jeffrey Seow, Nayung Lee, Annachiara
Rosa, Chloe Roustan, Catherine F. Houlihan, Rogier W. Sanders, Ravindra K.
Gupta, Peter Cherepanov, Hans J. Stauss, Eleni Nastouli, on behalf of the SAFER
Investigators, Katie J. Doores, Marit J. van Gils, and Laura E. McCoy
Supplemental Information

Figure S1. Substitution in SARS-CoV-2 based on SARS-CoV sequence, related to Figure 1.

Figure S2. Binding and neutralization titers across two infection cohorts, related to Figure 3.
Figure S1. Substitution in SARS-CoV-2 based on SARS-CoV sequence, related to Figure 1.

(A) Alignment of RBD from pseudotype SARS-CoV-2 Spike plasmid with SARS-CoV consensus amino acids. Sequences were aligned using
the Clustal Omega (1.2.4) tool. The amino acid positions removed in the Spike plasmid are highlighted in grey and the residues of SARS-CoV
that replaced them are indicated in red text. Amino acid positions for the first residue changed in each individual mutant is positioned above the
first mutated residue in any given mutant. A ¶ by the residue number indicates the pseudotyped virus carrying this substitution did not yield
sufficiently infectious titre for analysis of antibody sensitivity. The RBM region in both sequences is underlined. Conservation symbols are listed
below the aligned residues and are as follows: an * (asterisk) indicates positions that have a single, fully conserved residue; a : (colon)
indicates conservation between groups of strongly similar properties; a . (period) indicates conservation between groups of weakly similar
properties; a (gap) indicates no conservation.
(B) SARS CoV-2 RBD (PDB code 6VXX) showing WT (top) and mutated (bottom) amino acid side chains at residues of interest. Using UCSF
Chimera, amino acid side chains were altered at select positions using Dunbrack 2010 backbone-dependent rotamer library (Shapovalov &
Dunbrack, 2011). Note that the image does not show structure alterations that may occur as a result of mutated residues.
(C) Each SARS-CoV-2 pseudotype mutant is listed on the left and the mAb clusters are shown on the right. Mutations that result in a complete
loss of function are indicated by a line joining the mutant to the mAb. Mutations that result in a partial loss of function are indicated by a dashed
line joining the mutant to the mAb. Two mutations that had minimal/no effect on function are not linked to any mAb.

Shapovalov, M. V., & Dunbrack, R. L., Jr. (2011). A smoothed backbone-dependent rotamer library for proteins derived from adaptive kernel
density estimates and regressions. Structure, 19(6), 844-858. doi:10.1016/j.str.2011.03.019
A
324 346
pcDNA-SARS-CoV2 Spike SETKCTLKSFTVEKGIYQTSNFRVQPTESIVRFPNITNLCPFGEVFNATRFASVYAWNRK
SARS-CoV-consensus AELKCSVKSFEIDKGIYQTSNFRVVPSGDVVRFPNITNLCPFGEVFNATKFPSVYAWERK
:* **::*** ::*********** *: .:*******************:* *****:**
373 384
pcDNA-SARS-CoV2 Spike RISNCVADYSVLYNSASFSTFKCYGVSPTKLNDLCFTNVYADSFVIRGDEVRQIAPGQTG
SARS-CoV-consensus KISNCVADYSVLYNSTFFSTFKCYGVSATKLNDLCFSNVYADSFVVKGDDVRQIAPGQTG
:**************: ********** ********:********::**:**********
417 439¶ 444 452 455 459 470 474¶
pcDNA-SARS-CoV2 Spike KIADYNYKLPDDFTGCVIAWNSNNLDSKVGGNYNYLYRLFRKSNLKPFERDISTEIYQAG
SARS-CoV-consensus VIADYNYKLPDDFMGCVLAWNTRNIDATSTGNYNYKYRYLRHGKLRPFERDISNVPFSPD
************ ***:***:.*:*:. ***** ** :*:.:*:*******. :. .
477 481¶ 494
pcDNA-SARS-CoV2 Spike STPCNGVEGFNCYFPLQSYGFQPTNGVGYQPYRVVVLSFELLHAPATVCGPKKSTNLVKN
SARS-CoV-consensus GKPCTP-PALNCYWPLNDYGFYTTTGIGYQPYRVVVLSFELLNAPATVCGPKLSTDLIKN
..**. .:***:**:.*** *.*:***************:********* **:*:**

B WT SARS-CoV-2 RBD
C
ES324GD
ES324 COVA2-29
S373F Cluster I
KVG444-6
COVA1-18
TEI470-2 S373 P384A
L452
S494
P384 COVA2-07
K417V
Cluster III

KVG444-446TST COVA1-16
LF455 K417

L452K COVA1-12 Cluster VI

Mutant SARS-CoV-2 RBD

TEI470-2NVP LF455YL COVA2-02 Cluster VII

KVG444-6TST
L452K
S494D TEI470-472NVP COVA2-17
ES324GD Cluster IX
S373F
S494D
LF455YL COVA1-21 Cluster XI
P384A
Partial abrogation
K417V Complete abrogation
Figure S2. Binding and neutralization titers across two infection cohorts, related to Figure 3.

Binding and neutralization titers were generated as described in the Methods. Concentrations of S1-specific serum IgG (pg) at ID50 dilutions
were calculated using the IgG titers quantified via the semi-quantitative ELISA and the known ID50 value. Only sera that gave a measurable titer
in both semi-quantitative ELISA and pseudotype neutralization assay were included. Column statistics below were generated in Prism
Graphpad and the data were analyzed by a non-parametric Mann-Whitney U test.
 Binding titer Neutralization titer Specific IgG
(S1 IgG µg/ml) (ID50) at ID50
Mild Severe Mild Severe Mild Severe
n= 105 94 99 93 92 91
Minimum 0.3 2 50 50 12.86 68.06
25% Percentile 1.6 17 201.3 513.8 253.9 971.3
Median 3.9 46.5 342.3 1466 558.6 1757
75% Percentile 7.2 122.8 666.4 2556 1155 3073
Maximum 68 529 4152 20601 2907 45053

Mean 5.562 89.56 603.8 2347 736.5 3181

Std. Deviation 7.429 108.2 718.5 3008 604.7 5540
Std. Error of Mean 0.725 11.16 72.21 311.9 63.04 580.8

Lower 95% CI of mean 4.124 67.4 460.5 1727 611.3 2027

Upper 95% CI of mean 6.999 111.7 747.1 2966 861.7 4335

Mann Whitney test Mann Whitney test Mann Whitney test

P value < 0.0001 < 0.0001 < 0.0001
Exact or approximate P value? Exact Exact Exact
P value summary **** **** ****
Significant (P < 0.05) Yes Yes Yes
One- or two-tailed P value? Two-tailed Two-tailed Two-tailed
Sum of ranks in column A,B 6217, 13684 6862, 11667 5777, 11059
Mann-Whitney U 651.5 1912 1499